Assumptions in 3D-QSAR

The binding of a ligand (e.g., a compound in its bioactive conformation) to a macromolecule (e.g., receptor) can be, at equilibrium, expressed as  

Besides neglecting solvation and entropic terms, other limitations in the CoMFA description of the ligand—receptor interaction include (1) the lack of hydrophobic terms, which can be explicitly included using or the 

In this section, we present a brief description of some current 3D-QSAR methods. For a more complete description of the philosophies and inner workings of these applications, the reader is referred to the books edited by Kubinyi, Sanz, and van de Waterbeemd, and the references cited in these works. 

Perhaps the forerunner of all three-dimensional, grid-based QSAR techniques was DYLOMMS, developed between 1983 and 1987.- Whereas many modifications have since been incorporated into the CoMFA method,- the basic philosophy has not changed. In essence, CoMFA does exactly what its name implies—it compares the molecular potential energy fields (by default steric and electrostatic) of a series of molecules and searches for differences and similarities that can be correlated with differences and similarities in target property values. Recently reviewed by Cramer and co-workers, this method is detailed later in this chapter. 

GRID provides the user with a wide variety of atomic probe types (like CoMFA) and funaional group probe types (a feature not available in CoMFA). GOLPE, as originally presented, used a variable selection technique based on 

---

ASSUMPTIONS IN 3D-QSAR. THE BIOACTIVE CONFORMATION AND BIOLOGICAL ACTIVITY... 

---